Re: [H] EIDE-speak?

2007-12-25 Thread DHSinclair

replies inline...
At 20:37 12/24/2007 -0800, Tharin O. wrote:
Some of those drives had a jumper to force the 504mb drive geometry for 
compatibility. I imagine you would have spotted this though because it is 
normally on the drive label.


Yes, the two 'lower' members of this drive's family to have this feature. 
However, per the spec manual the dpea-31080 has this 504mb geometry feature 
disabled. Says that jumper position 7 (pins 13-14) is ignored. Hmm. I 
wonder :)


Is it actually sharing the ide cable with another drive or is it by 
itself? Sometimes the older drives had to be jumpered for Single instead 
of Master.


Just now the hd is on the Master connector (set for CableSelect) and the 
CDROM is on the Slave connector (set for CableSelect).  The next test will 
be to remove the 'CableSelect' option and promote each device to its' own 
position.  Suspect I might have been testing whether CS works or 
not.  Perhaps not with this bios.. :)



Obviously a 440BX chipset board should handle a 1gb drive. I think that 
one was new enough to go up to the 32gb barrier or possibly the 137gb barrier.


Seems logical, but back in 1998 when this m/b was new, IIRC, a 40GB had was 
a really big deal. And a 137GB hd was just a dream!  My how time flys. And 
technology marches on.


If you enter CMOS utility and use the Auto Detect IDE option does it 
display the values it is detecting for the sector, head, cylinder, disk 
type, etc. ?? Is it LBA, CHS, or LARGE mode?


Now this gets to most for my present confusion. For SG's I did use the 
Auto Detect IDE option. What I received for the hd was/is a bit queer. I 
get the small read window with:

Select Primary Master   Option (N=Skip) : []
Options SizeCylsHeadPreComp LandZ   Sector  Mode
2(Y)1083525 64  0   209963  LBA
1   1083210016  65535   209963  NORMAL
3   1083525 64  65535   209963  LARGE

Obviously, I choose the one that matches the hd's spec'd geometry. That 
would be (1).  Confused by why choice #2 has (Y) appended.  Also notice the 
even though this detect routine seems to acknowledge the CDROM, when I go 
back to the base cmos screen, the primary slave line indicates that there 
is nothing present.  Not quite Bios-Hell, but close.


 What happens if you force LBA, CHS, or LARGE and leave the geometry 
details set to Auto?


This I will try in a bit. Except I think that if I leave the details set to 
Auto, I can not force any of the values. From what I see, if I choose Auto, 
bios rules.


Go into FDISK and look under NON-DOS Partitions (#4 i think)?, anything 
there? Try using FDISK /MBR at the command prompt of your DOS boot disk 
to rewrite the master boot record.


Yes, this I will try also. Though, in all these years I have never been 
able to get this command to work. Will try again.



It doesn't really matter I suppose, but I'm having trouble understanding 
why are  you dealing with DOS and trying to get a pretty ESCD table in 
the first place?


Perhaps just my own stupid pig-headedness. I grew up in computers learning 
that it was best if devices in the ESCD table had their own IRQs to use. 
And, way back then, some devices liked some IRQs better than others.  Now, 
I suppose this is just so much folklore and habitual ritual. Nothing more.



I thought you were gonna run FreeBSD or Linux??


Yes, that is what I will do eventually.  The first install try got goofed 
up by me. It started booting with more that a few command line screens that 
indicated something went badly wrong.  I am starting over. And, part of it 
is finding the hd dorked up partition-wise, install-wise, pebcak-wise. :)
When all else fails, I retreat to the last known OS I could sorta 
drive.  That be MS-DOS.


I'm drawing upon neurons that were formed during late grade school and jr. 
high. Not quite sure if that makes me feel young or old. ;)


We are as young and old as we feel. Put the neurons to nap mode. I mean you 
no harm with this. Have another EggNog! Merry Christmas!  This is goof-off 
project for me. If it does not work, well, my test of FreeNAS just ends and 
I'll find something else.  Or, maybe just buy some newer 
hardware :)

Best,
Duncan




-Tharin Olsen


DHSinclair [EMAIL PROTECTED] wrote:
MS-DOS does not seen the hd any differently. Still 504MB.  Have not tested
the CDROM on the same cable yet. Still Playing..
I just think that DOS does not care for a 1083MB hard drive from IBM!  I
could be real wrong here too!




[H] EIDE-speak?

2007-12-24 Thread DHSinclair

Am I correct:

Old EIDE: Black connector (m/b) - Black connector (master) - Black 
connector (slave)


Pata EIDE: Blue connector (m/b) - Gray connector (master) - Black connector 
(slave)


Still wondering about the plugged pin in the m/b connector.
Have lots of these.  Can not use them.  My m/b has a pin where it should 
not be.

Best,
Duncan



Re: [H] EIDE-speak?

2007-12-24 Thread Tharin Olsen
All IDE devices/cables are parallel ata. The Blue/gray/black cables you have 
should be the ATA 66/100 rated cables with 80 conductors/wires. The older style 
only had 40 but was only good up to ATA 33.  Black (far end of cable) is master 
and gray (middle connector) is slave. You are right about Blue being for the 
host controller.

Pin 20 should be the one that is plugged on your cables. You might be able to 
remove it with a pick if it isn't actually molded that way. Newer drives and 
motherboards have that pin removed and the cables are keyed as such.  It was 
also like that on early compaqs, dells, etc. I remember having lots of pulled 
cables from those kind of systems and not being able to use them on whitebox 
motherboards or drives because they weren't keyed. The Tier1's seemed to like 
have their cables keyed, I guess they couldn't put stuff on backwards at the 
factory that way. 

-Tharin Olsen

DHSinclair [EMAIL PROTECTED] wrote: Am I correct:

Old EIDE: Black connector (m/b) - Black connector (master) - Black 
connector (slave)

Pata EIDE: Blue connector (m/b) - Gray connector (master) - Black connector 
(slave)

Still wondering about the plugged pin in the m/b connector.
Have lots of these.  Can not use them.  My m/b has a pin where it should 
not be.
Best,
Duncan




Re: [H] EIDE-speak?

2007-12-24 Thread DHSinclair

Thanks Tharin O.,
inline below..
At 16:56 12/24/2007 -0800, you wrote:
All IDE devices/cables are parallel ata. The Blue/gray/black cables you 
have should be the ATA 66/100 rated cables with 80 conductors/wires. The 
older style only had 40 but was only good up to ATA 33.  Black (far end of 
cable) is master and gray (middle connector) is slave. You are right about 
Blue being for the host controller.


Yes, all my research today gets right to this. The funny part is the m/b 
(bios) does not seem to recognize the hd or the CDROM, though it does seem 
to use them. Depending on what I ask to be done.  Hmm!




Pin 20 should be the one that is plugged on your cables.


Yes, pin 20 is plugged in all the SPARE EIDE cables I own. The problem is 
that pin-20 is NOT missing on my m/b!  Pin-20 is clearly sticking up and 
staring at me.  OK.  I am now testing a modified EIDE cable.  I cut out, 
dug out sorta, well, trashed a brand new (?) EIDE cable to test this pin-20 
business...
MS-DOS does not seen the hd any differently. Still 504MB.  Have not tested 
the CDROM on the same cable yet. Still Playing..
I just think that DOS does not care for a 1083MB hard drive from IBM!  I 
could be real wrong here too!



You might be able to remove it with a pick if it isn't actually molded 
that way. Newer drives and motherboards have that pin removed and the 
cables are keyed as such.  It was also like that on early compaqs, dells, 
etc. I remember having lots of pulled cables from those kind of systems 
and not being able to use them on whitebox motherboards or drives because 
they weren't keyed. The Tier1's seemed to like have their cables keyed, I 
guess they couldn't put stuff on backwards at the factory that way.


Been there, done that.  Still playing.  No not a Compaq, not a Dell, not an 
HP.  I built this turkey all by myself back in 1999.  Not a bad thing here. 
Just some odd stuff I see and do not expect.

Best/Merry Christmas/Noel/Mele Kalikimaka/Feliz Navidad/
Duncan


-Tharin Olsen

DHSinclair [EMAIL PROTECTED] wrote: Am I correct:

Old EIDE: Black connector (m/b) - Black connector (master) - Black
connector (slave)

Pata EIDE: Blue connector (m/b) - Gray connector (master) - Black connector
(slave)

Still wondering about the plugged pin in the m/b connector.
Have lots of these.  Can not use them.  My m/b has a pin where it should
not be.
Best,
Duncan




Re: [H] EIDE-speak?

2007-12-24 Thread Tharin Olsen
Some of those drives had a jumper to force the 504mb drive geometry for 
compatibility. I imagine you would have spotted this though because it is 
normally on the drive label. Is it actually sharing the ide cable with another 
drive or is it by itself? Sometimes the older drives had to be jumpered for 
Single instead of Master.

Obviously a 440BX chipset board should handle a 1gb drive. I think that one was 
new enough to go up to the 32gb barrier or possibly the 137gb barrier. If you 
enter CMOS utility and use the Auto Detect IDE option does it display the 
values it is detecting for the sector, head, cylinder, disk type, etc. ?? Is it 
LBA, CHS, or LARGE mode? What happens if you force LBA, CHS, or LARGE and leave 
the geometry details set to Auto? Go into FDISK and look under NON-DOS 
Partitions (#4 i think)?, anything there? Try using FDISK /MBR at the command 
prompt of your DOS boot disk to rewrite the master boot record.

It doesn't really matter I suppose, but I'm having trouble understanding why 
are  you dealing with DOS and trying to get a pretty ESCD table in the first 
place? I thought you were gonna run FreeBSD or Linux?? I'm drawing upon neurons 
that were formed during late grade school and jr. high. Not quite sure if that 
makes me feel young or old. ;)

-Tharin Olsen


DHSinclair [EMAIL PROTECTED] wrote:
MS-DOS does not seen the hd any differently. Still 504MB.  Have not tested 
the CDROM on the same cable yet. Still Playing..
I just think that DOS does not care for a 1083MB hard drive from IBM!  I 
could be real wrong here too!